Introduction

The ecological importance of large gelatinous carnivo-rous zooplankton such as cnidarians and ctenophores hasbeen increasingly recognized, since their unusual aggrega-tions and population increases have been reported fre-quently in many parts of the world’s ocean in recentdecades (Arai 2001, Mills 2001, Brodeur et al. 2002, Pur-cell 2005, Kawahara et al. 2006, Purcell et al. 2007). Previ-ous studies suggest that the abundance of such jellyfish canbe a key factor in regulating the trophic structure of marineplankton communities; when they occur abundantly, theirpredation impact may be so intensive as to affect the popu-lation size and species composition of the plankton commu-nities (Lindahl & Hernroth 1983, Olesen 1995, Schneider& Behrends 1998, Pagès et al. 2001). At the same time, jel-lyfish also reduce fish standing stocks, either by direct pre-dation on fish eggs and larvae or by competing with plank-tivorous fish and fish larvae for available zooplankton prey(Möller 1980, Purcell & Arai, 2001, Hansson et al. 2005).In addition, jellyfish hamper fishing activity by clogging

and bursting fishing nets, and cause problems to coastalpower plants by blocking cooling water intakes (Kuwabaraet al. 1969, Rajagopal et al. 1989, Purcell et al. 2007).

In Japanese coastal waters, like many other coastal wa-ters, the moon jelly Aurelia aurita s.l. Linnaeus is the mostcommon and abundant scyphozoan species (see Dawson2003, Dawson & Martin 2001 for sibling species), and it isparticularly abundant in eutrophic waters such as TokyoBay and the Inland Sea of Japan. In Tokyo Bay, the massoccurrence of A. aurita medusae began in the 1960s, whenassociated problems of clogged screens of power plant sea-water intakes were reported (Kuwabara et al. 1969, Yasuda,1983). Since then, A. aurita has become the most dominantspecies in the zooplankton community (Omori et al. 1995)and exerts high predation pressure, equivalent to 5 to 162%of mesozooplankton biomass per day in spring and summer(Kinoshita et al. 2006). In the Inland Sea of Japan, the in-crease in A. aurita population became significant in the1990s, when both increasing water temperature, particularlyin winter, and decreasing of zooplanktivorous fish popula-tions were prominent (Uye et al. 2003, Uye & Ueta 2004).In Ondo Straight, a central part of the Inland Sea of Japan,the average biomass of A. aurita (66.0 mg C m�3) was much

Seasonal variations in the trophic relationship between thescyphomedusa Aurelia aurita s.l. and mesozooplankton ina eutrophic brackish-water lake, Japan

CHANG-HOON HAN, MASATO KAWAHARA & SHIN-ICHI UYE*

Graduate School of Biosphere Science, Hiroshima University, 4–4 Kagamiyama 1 Chome, Higashi-Hiroshima 739–8528, Japan

Received 4 July 2008; Accepted 29 October 2008

Abstract: The seasonal variations in trophic relationship between the moon jelly Aurelia aurita s.l. and mesozoo-plankton were investigated in a brackish-water lake, Honjo District, Japan from June 2005 to August 2006. Themedusae occurred abundantly (average abundance and biomass: 0.55 medusae m�3 and 58.8 mg C m�3, respectively)during warm seasons (i.e. June–November, 2005), but were very scarce or absent during the remaining seasons. Themesozooplankton biomass fluctuated from 1.3 to 150 mg C m�3 (overall average: 60.5 mg C m�3) irrespective of themedusa biomass variation. Mesozooplankton were preyed upon by medusae almost non-selectively; the small copepodOithona davisae and bivalve larvae were the predominant prey, comprising 52–99% (average: 85%) of the gastricpouch contents. The medusa population ingestion rate on mesozooplankton varied from 0.11 to 12.8 mg C m�3 d�1,which corresponded to 0.6 to 29% of the mesozooplankton biomass per day and to 1.6 to 47% of mesozooplanktondaily production rate. A. aurita medusae were certainly a key component of the zooplankton community, but they didnot exert any significant top-down control as to suppress mesozooplankton biomass in this eutrophic lake.

Key words: Aurelia aurita, Japan, mesozooplankton, predation impact, seasonal occurrence

Plankton Benthos Res 4(1): 14–22, 2009

Corresponding author: Shin-ichi Uye; E-mail, suye@hiroshima-u.ac.jp

Plankton & Benthos Research

© The Plankton Society of Japan

higher than the micro- and mesozooplankton biomass (av-erage: 23.7 mg C m�3) and its predation impact was equiva-lent to 26% of zooplankton biomass per day from May toAugust (Uye & Shimauchi 2005).

Honjo District is an enclosed shallow brackish-water lake(area: 16.2 km2, average depth: 5.1 m) in the northern partof Lake Nakaumi, Shimane Prefecture (Fig. 1). The HonjoDistrict was separated from other waters by construction ofa bank in 1981, but with connections via two narrow inletsso that the water exchange between the Honjo District andthe Lake Nakaumi is limited (daily exchange rate: 2.26% ofwater volume of the Honjo District, Koike et al. 1999). Fre-quent mass occurrences of A. aurita medusae during sum-mer were reported by local fishermen after the embank-ment, but no research has been conducted on their seasonalpopulation dynamics in the Honjo District. Meanwhile, thiseutrophic lake harbors one of the world’s highest mesozoo-plankton biomasses (annual average: 71 mg C m�3, Uye etal. 2004), and hence it is our prime aim to understand theprey and predator relationship between mesozooplanktonand A. aurita medusae. We investigated the seasonal varia-tions in abundance, biomass, gut contents and populationingestion rate of A. aurita medusae, together with mesozoo-plankton abundance, taxonomic composition, biomass andproduction rate.

Materials and Methods

Aurelia aurita medusa abundance, biomass and gut con-tents

Prior to the initiation of our regular samplings, we con-ducted a sighting survey on the geographical distribution ofA. aurita medusae by running a boat along 3 cross transectsof the Honjo District. Since we encountered aggregatedmedusae most often in the central part of the lake and very

few near the shore, we set up a sampling area (approximatediameter: 3 km) in the central Honjo District (Fig. 1) duringour investigation (at ca. 2-week intervals) from June 2005to August 2006. On each sampling occasion, surface andbottom (depth: ca. 6 m) water temperatures, salinities anddissolved oxygen concentrations were measured by a multi-ple water checker (Horiba, U-20). To lessen the samplingvariance due to heterogeneous distribution (see Results),medusae were caught by 5 to 7 horizontal tows of a net(0.6 m mouth diameter, 2.0 m length, 505 mm mesh size andfitted with Rigosha flowmeter) at 1–3 m depth at 1.0–1.5 ms�1 for 3–5 minutes. Total wet weight and numbers ofmedusae caught by each tow were determined, and the belldiameter and wet weight were measured for specimenscaught during separate tows.

Wet, dry and carbon weights were determined formedusae collected in August 2005. Medusae were trans-ported alive to the laboratory in 20 L containers. After rins-ing with distilled water and blotting, wet weight was mea-sured. Dry weight was then measured after drying in anelectric oven at 60–65°C for 4–6 days until constant weightwas reached. Carbon contents of 15–22 mg aliquots of pul-verized dried medusa were determined using a CHN ana-lyzer (Yanagimoto, MT-3) with antipyrine as the standard.

On each sampling date, 5 to 10 medusae were collectedby a scoop net, and were individually preserved in 10% for-malin lake-water solution. Within several days of sampling,the gastric pouch was dissected and the food contents re-moved from the gastric filaments by using a jet of tap water.The gut contents in the wash were retained with a 40 mmsieve and examined under a dissecting microscope. Appro-priate body length of zooplankters (e.g. prosome length forcopepod copepodites, body length for copepod nauplii,cladocerans and polychaete larvae, body width for bivalvelarvae, see Uye 1982) were measured with a video microm-

Trophic relationship between common jellyfish Aurelia aurita and mesozooplankton 15

Fig. 1. Sampling site of Aurelia aurita medusae and mesozooplankton (shaded) in Honjo Disrict, Shimane Prefecture, Japan.

eter (Mitani, WinRoof) and automatically converted to car-bon weights using predetermined length-weight regressions(see Uye & Shimazu 1997 for details).

Mesozooplankton abundance, biomass and productionrate

Mesozooplankton samples were collected by verticalhauls of a plankton net (25 cm mouth diameter, 80 mmmesh size) from the bottom to the surface and immediatelyfixed with 5% formalin lake-water solution. Because of therelatively homogeneous distribution of mesozooplankton inthe Honjo District (overall average of variance of theirabundance among 7 samples: 22%) found in a previousstudy (24 sampling occasions for two years, Uye et al.2004), a single sample was examined on each occasion.The sample was split into 1/16–1/32 subsamples with aMotoda box splitter, from which at least 200 specimenswere identified and counted under a dissecting microscope.Their appropriate body lengths were measured and con-verted to carbon weights, as described above. The produc-tion rate of each taxonomic group (P, mg C m�3 d�1) wasestimated based on its biomass (B, mg C m�3) and experi-mentally determined specific growth rates (g, d�1) (seeTable 1 of Uye & Shimazu 1997):

P�B�g .

Summation of this for all taxonomic groups gives themesozooplankton community production rate (CP,mg C m�3 d�1):

CP�SP .

Feeding selectivity and predation impactThe feeding selectivity of A. aurita was determined

based on the Chi-square (c 2) method comprising of two�two configured comparisons between the average numberof each prey taxon in the gastric pouch and correspondingnumber in the ambient water (see Pearre 1982 for details)as has been commonly used in many jellyfish prey selectiv-ity studies (Graham & Kroutil 2001, Purcell & Sturdevant2001, Sullivan et al. 1994). Selectivity, C, is given by:

C��(c 2/n)1/2

where n is the total number of a given prey taxon both inthe medusa and in the water.

The A. aurita population ingestion rate on mesozoo-plankton was estimated by:

PIR�24�GC�DT�1�N

where PIR is the population ingestion rate (mg C m�3 d�1),GC is the average carbon weight of gut contents(mg C medusa�1), N is medusa abundance (medusae m�3)and DT is digestion time (h). The digestion time was notmeasured but assumed to be 1.0 h according to the previousstudies by Ishii & Tanaka (2001) and Uye & Shimauchi(2005), because of similar prey taxa (i.e. Oithona davisaeFerrari & Orsi, Acartia omorii Bradford, bivalve larvae and

16 C. HAN, M. KAWAHARA & S. UYE

Tab

le1.

Sum

mar

y of

dai

ly p

reda

tion

impa

ct o

f A

urel

iam

edus

a po

pula

tion

s on

pre

y zo

opla

nkto

n bi

omas

s an

d pr

oduc

tion

rat

es.

Loc

atio

n an

d se

ason

Med

usa

abun

danc

eM

edus

a P

rey

type

Inge

stio

n ra

teD

aily

pre

dato

ry im

pact

Ref

eren

ceor

bio

mas

sbe

ll d

iam

eter

or c

lear

ance

rat

eon

zoo

plan

kton

Kie

l Big

ht, G

erm

any

0.00

2–0.

16 m

ed.m

�3

20cm

Mix

ed z

oopl

ankt

on40

–80

mg

C m

ed.�

1d�

164

% o

f pr

oduc

tion

rat

eS

chne

ider

& B

ehre

nds

Sum

mer

2–44

mg

Cm

�3

(199

4)B

ornh

olm

Bas

in, G

erm

any

�0.

023

med

.m�

37–

15cm

Cop

epod

s &

cla

doce

rans

�38

5 pr

ey m

ed.�

1d�

1�

0–1.

2% o

f co

pepo

d bi

omas

sB

arz

& H

irch

e (2

005)

Spr

ing-

fall

�4,

815

prey

med

.�1d�

10.

1–7.

9% o

f cl

adoc

eran

bio

mas

sK

erti

nge

Nor

, Den

mar

k2–

248

med

.m�

32–

6cm

Rot

ifer

s0.

01–2

.4l�

1m

ed.�

1h�

1�

0–35

1% o

f bi

omas

sO

lese

n (1

995)

Spr

ing-

fall

Pri

nce

Wil

liam

Sou

nd, U

SA

�0.

003

med

.m�

37–

26cm

Cop

epod

s &

cla

doce

rans

506

prey

med

.�1d�

10.

05%

of

cope

pod

biom

ass

Pur

cell

(20

03)

Sum

mer

634

prey

med

.�1d�

12.

3% o

f cl

adoc

eran

bio

mas

sIn

land

Sea

of

Japa

n, J

apan

65m

gC

m�

321

cmM

ixed

zoo

plan

kton

6.07

mg

Cm

ed.�

1d�

126

% o

f bi

omas

sU

ye &

Shi

mau

chi (

2005

)S

umm

erTo

kyo

Bay

, Jap

an18

.3–2

36m

gC

m�

3 *13

–18

cmM

ixed

zoo

plan

kton

0.16

l�1gW

W�

1h�

15–

162%

of

biom

ass

Kin

oshi

ta e

t al.

(200

6)S

prin

g-su

mm

erH

onjo

Dis

tric

t, Ja

pan

0.6–

1.3

med

.m�

311

–19

cmM

ixed

zoo

plan

kton

0.06

–29.

8m

gC

med

.�1d�

10.

6–29

% o

f bi

omas

sP

rese

nt s

tudy

Sum

mer

-fal

l20

.6–1

20m

gC

m�

31.

6–47

% o

f pr

oduc

tion

rat

e

*C

onve

rted

fro

m a

vera

ge d

epth

s (2

4 an

d 44

m)

of s

tati

ons

in T

okyo

Bay

.

gastropod larvae), medusa bell diameter: (range: 10–25 cm)and temperature (range: 18–22°C) in these studies.

Results

Environmental parametersTemperature fluctuated greatly from 3.2 to 30.7°C at the

surface and from 3.2 to 28.3°C at the bottom (Fig. 2a).Salinity varied from 12 to 24 at the surface and from 14 to25 at the bottom with overall averages of 19.8 and 20.8, re-spectively (Fig. 2b). Dissolved oxygen concentration wasusually saturated (�7.5 mg O2 L�1) at the surface, but hy-poxic at the bottom (minimum: 0.5 mg O2 L�1) during sum-mer (Fig. 2c).

Mesozooplankton abundance, biomass and productionrate

Mesozooplankton abundance showed considerable fluc-tuations with season, occurring more abundantly from Juneto October (average: 23.4�104 ind. m�3) than in the remain-ing periods (average: 4.4�104 ind. m�3) (Fig. 3a). The an-

nual peak abundance was recorded on 5 August 2005(44.4�104 ind. m�3) and on 12 July 2006 (73.3�104 ind. m�3).

In terms of biomass, mesozooplankton also fluctuatedmarkedly from 1.3 to 150 mg C m�3 over the study period(Fig 3b). Largely according to the seasonal numerical varia-tion, the biomass was also high from June to October (aver-age: 70 mg C m�3) with prominent peaks on 5 July (116mg C m�3) and on 16 October (150 mg C m�3) in 2005 andon 20 August (128 mg C m�3) in 2006. These biomasspeaks were composed primarily (65–99%) of the smallcopepod Oithona davisae and bivalve larvae (Fig. 3d).

Trophic relationship between common jellyfish Aurelia aurita and mesozooplankton 17

Fig. 2. The seasonal variations in surface and bottom (6 m) tem-perature (a), salinity (b) and dissolved oxygen (c) concentration inHonjo District.

Fig. 3. The seasonal variations in abundance (a), biomass (b),production rate (c) and biomass-based taxonomic composition ofmesozooplankton (d) in Honjo District.

Mesozooplankton production rate varied from 2.2 to260 mg C m�3 d�1 (Fig. 3c). As a result of positive tempera-ture effects on the specific growth rate, the production ratetended to be higher relative to biomass during warm sea-sons. An extremely high production rate on 16 October2005 was attributed to the dominance of small-sized bi-valve larvae.

Aurelia aurita medusa abundance, biomass and gut con-tents

Measurements of 600 A. aurita specimens from theHonjo District gave the following relationship between wetweight (WW, g) and bell diameter (BD, cm) (Fig. 4):

WW�0.089BD 2.70 .

Dry weight and carbon contents were measured for 18medusae, with wet weights ranging from 13 to 78 g. Aver-age dry weight was 2.6% (SD�1.1%) of wet weight, andaverage carbon weight was 4.8% (SD�2.3%) of dryweight. There was no significant size-dependent differencein these relative weights. Hence, the average wet : dry :carbon weight ratio for A. aurita medusae from the HonjoDistrict was 100 : 2.6 : 0.13, which was used for convertingwet weight biomass to carbon biomass.

Although we repeated 5–7 net tows at randomly selectedlocations in the central Honjo District to alleviate the sam-pling variance (overall average: 60%), the sample size ofeach tow differed greatly, particularly in summer of 2005when medusae were often in patchy aggregations. In an ex-treme case on 5 August 2005, one towed sample contained193 medusae but the next one contained only 2 medusae.Hence, the sampling variance among 7 tows on this datewas as large as 144%.

The abundance of A. aurita medusae increased from0.29 medusae m�3 in June, peaked on 20 August (1.33medusae m�3) and then suddenly declined on 8 September2005 (Fig 5a). A strong typhoon, No. 14 (maximum windspeed in Shimane Prefecture: 97 km h�1), passed on 6 Sep-tember. Although only a few medusae were caught by ourregular net samplings on 8 September, our scuba diving ob-servations revealed that a large number of medusae wereconcentrated in the well-aerated bottom layer (Fig. 2c).After this sudden decline, the abundance recovered to0.59 medusae m�3 in October and then disappeared untilresurgence in June 2006 (Fig. 5a). The medusa populationwas much smaller in 2006 (maximum abundance: 0.12medusae m�3) than in 2005 (Fig. 5a).

In 2005, the size of medusae was smallest in June (aver-

18 C. HAN, M. KAWAHARA & S. UYE

Fig. 4. Relationship between bell diameter and wet weight ofAurelia aurita caught in Honjo District.

Fig. 5. The seasonal variations in abundance (a), bell diameter(b), carbon weight (c) and biomass (d) of Aurelia aurita in HonjoDistirct. Vertical lines denote SD. Open circles indicate zero.

age bell diameter and carbon weight: 10.5 cm and 20.6 mgC, respectively), and gradually increased until November,when average bell diameter and carbon weight was 14.3 cmand 153 mg C, respectively (Fig. 5b, c). There were no sig-nificant changes in the frequency distributions of medusabell diameter in 2005 and individual cohorts could not beidentified. The growth of medusae was, however, clearlytraced during June and July 2006, when the average bell di-ameter and carbon weight increased from 10.4 to 17.1 cmand from 60.3 to 278 mg C, respectively.

The pattern of seasonal variation in medusa biomass wassimilar to that of abundance; increasing from 20.6mg C m�3 in June to 120 mg C m�3 in August then decreas-ing to 35.7 mg C m�3 in November (Fig. 5d). The averagebiomass between June and November was 58.8 mg C m�3.The average medusa biomass in June and July 2006 was10.0 mg C m�3.

The daily ingestion rate varied markedly from 0.06 to29.8 mg C medusa�1 d�1 in 2005 (Fig. 6a). The highest in-gestion rate was recorded on 16 October 2005, when preybiomass and production rate were also highest (Fig. 3a).

The gut content examination revealed that A. auritamedusae ingested all the mesozooplankton taxa present inthe Honjo District. The small copepod O. davisae and bi-valve larvae were the most important prey items (Fig. 6b),accounting for 52 to 99% (average: 85%) of the gut con-

Trophic relationship between common jellyfish Aurelia aurita and mesozooplankton 19

Fig. 6. The seasonal variations in ingestion rate (a) and carbonbiomass composition of gut contents (b) of Aurelia aurita inHonjo District. Vertical lines donote SD.

Fig. 7. The seasonal variations in feeding selectivity of Aureliaaurita estimated from gut contents and ambient mesozooplanktonin Honjo District.

Fig. 8. The seasonal variations in population ingestion rate (a),predation impact on mesozooplankton biomass (b) and predationimpact on mesozooplankton production rate (c) of Aurelia auritain Honjo District. Vertical lines denote SD.

tents, reflecting their dominance in the plankton. Othermajor prey taxa were copepods (Acartia hudsonica Pinheyand A. sinjiensis Mori) and gastropod larvae.

The feeding selectivity of medusae was determined for 5prey categories, i.e. A. hudsonica, A. sinjiensis, O. davisae,bivalve larvae and gastropod larvae. It ranged from �0.26for O. davisae to 0.25 for bivalve larvae, both recorded on13 November, but usually stayed near zero (Fig. 7), indicat-ing that A. aurita medusae fed on mesozooplankton preylargely non-selectively.

Aurelia aurita population impact on mesozooplanktonThe ingestion rate of the population of medusae on the

mesozooplankton community varied from 0.11 to 12.8mg C m�3 from June to November (Fig. 8a), which corre-sponded to 0.6 to 29% (average: 8.9%) of the mesozoo-plankton biomass per day (Fig. 8b), and to 1.6 to 47% (av-erage: 13.0%) of the mesozooplankton daily productionrate (Fig. 8c).

Discussion

The present study demonstrated the seasonality of the A.aurita medusa population in the Honjo District for the firsttime, where the population culminated during warm sea-sons (Fig. 5a), similar to that observed in other Japanesewaters such as Lake Hamana (Kuwabara 1969), UrazokoBay (Yasuda 1983), Tokyo Bay (Omori et al. 1995) and theInland Sea of Japan (Uye & Shimauchi 2005). In 2005,medusae already occurred before our survey started andthey seemed to grow relatively slowly to the maximum sizein autumn. In 2006, the resurgence of medusae occurred inJune after the disappearance of the population in the pre-ceding winter and spring. This indicates that a new medusageneration might start in spring, as has been commonly ob-served in other Japanese coastal waters (Yasuda 1969,Miyake et al. 1997, Watanabe & Ishii 2001). Unfortunately,the very scarce occurrence of ephyrae in our zooplanktonsamples failed to identify the precise timing of ephyra liber-ation from benthic polyps, which also remained undiscov-ered despite our scuba diving searches. Hence, it is still un-certain when and where medusa recruits come from for A.aurita population in the Honjo District.

Compared to 2005, the medusa population size wasmuch smaller in 2006, but vice versa for medusa body size(Fig. 5), indicating population density effect (Schneider &Behrends 1998). Such an annual variation in the occurrenceof medusae has been noticed by a local fisherman (S.Nakashima, personal communication), who has been oper-ating set-net fisheries in the Honjo District for �50 years.He also mentioned that medusae can persist during warmand mild winters and they tend to be more abundant in thefollowing summer. The 2005–2006 winter was very severe;average local air temperature in December 2005 was 4.0°C,while the corresponding average for the previous 10 yearswas 7.0°C (Meteorological Agency of Japan). An unusually

rapid decline of water temperature to the annual minimum(3.2°C in January 2006) might hamper the overwintering ofmedusae as well as recruitment of a new population.

Uye et al. (2004) reported that due to sufficient phyto-plankton food supply (average chlorophyll a concentration:4.7 mg L�1), the Honjo District is very productive for meso-zooplankton; the average biomass and production rate were71.0 mg C m�3 and 17.6 mg C m�3 d�1, respectively, duringa two-year survey from 1997 to 1999. The average meso-zooplankton biomass (60.5 mg C m�3) was similar to that ofUye et al. (2004) but the average production rate (32.4mg C m�3 d�1) was higher in our study. The predominanceof bivalve larvae in our plankton samples boosted themesozooplankton production rate, since small-sized bivalvelarvae have higher specific growth rates (g�0.4–1.0 d�1)than copepods (g�0.05–0.5 d�1). During the warm periodfrom June to November, 2005, the average medusa biomass(58.8 mg C m�3) was equivalent to the average mesozoo-plankton biomass (56.2 mg C m�3), demonstrating that A.aurita was certainly the most dominant species in the zoo-plankton community in the Honjo District.

Sullivan et al. (1994) reported that slow swimming preysuch as hydromedusae and barnacle larvae were more vul-nerable to predation by A. aurita. Graham & Kroutil (2001)also demonstrated that small copepods were more vulnera-ble than large a copepods (�1 mm) since the latter has ahigher escape ability. Zooplankton appearing in the HonjoDistrict almost entirely consists of meso- and microzoo-plankton (Godhantaraman & Uye 2003, Uye et al. 2004,present study) and hence they may be highly vulnerable. Infact, 5 major prey taxa (i.e. Oithona davisae, Acartia hud-sonica, A. sinjiensis, bivalve larvae and gastropod larvae)were ingested in an almost non-selective manner (Fig. 7).

The mesozooplankton ingestion rates by a medusa deter-mined in the Honjo District ranged from 0.06 to 28.9 mg Cmedusa�1 d�1, which were similar to those reported in pre-vious studies (i.e. 8–15 mg C medusa�1 d�1 in Kiel Bight,Schneider & Behrends 1994, 2.2–22.8 mg C medusa�1 d�1

in Tokyo Bay, Ishii & Tanaka 2001, 6.07 mg C medusa�1

d�1 in the Inland Sea of Japan, Uye & Shimauchi 2005).Previous studies estimated the predation impacts by Aureliamedusa populations on zooplankton communities in thefield, although different methods were employed dependingon the study; the results are summarized in Table 1. An ex-tremely high predation impact was reported by Olesen(1995) in Kertinge Nor, a brackish-water fjord in Denmark,where the A. aurita population ingested 351% of the rotiferbiomass per day. In Kiel Bight, Germany, the medusa pre-dation impact was equivalent to 64% of daily zooplanktonsecondary production (Schneider & Behrends 1994). InTokyo Bay, the predation pressure was estimated to varyfrom 5 to 162% of the zooplankton biomass per day (Ki-noshita et al. 2006). In the Inland Sea of Japan, it was esti-mated to be 26% of the zooplankton biomass per day (Uye& Shimauchi 2005). In contrast, the predation impacts re-ported by Purcell (2003) in Prince William Sound, Alaska,

20 C. HAN, M. KAWAHARA & S. UYE

and Barz & Hirche (2005) in Bomholm Basin, the BalticSea, were low (�ca. 8% of zooplankton biomass per day).In the Honjo District, the highest predation impact wasrecorded on 1 November 2005, when the A. aurita popula-tion (biomass: 81.4 mg C m�3) ingested 29% of the meso-zooplankton biomass (31.7 mg C m�3) per day and 47% ofthe daily production rate (19.8 mg C m�3 d�1). The corre-sponding average impacts between June and November2005 were 8.9 and 13.0%, respectively.

In conclusion, in a brackish-water lake called the HonjoDistrict, A. aurita medusae occurred only in warm seasons,and in terms of biomass this species was the most dominantcomponent of the zooplankton community. All major taxaof mesozooplankton were equally vulnerable to this gelati-nous predator, and the predation pressure by the medusapopulation was highest in autumn. However, no significanttop-down control by A. aurita was observed since the zoo-plankton biomass did not decrease concurrently. Mesozoo-plankton production potential might surpass the predationloss. This further suggests that A. aurita medusae might notbe food-limited but their seasonal population fluctuationmight be affected primarily by physical parameters such astemperature and deoxygenation in this eutrophic lake.

Acknowledgements

We thank Mr. Y. Mishiro for providing a motorboat forour sampling in the Honjo District and Miss M. Hayashi forhelp in zooplankton identification. This study was partiallysupported by a research grant from the Japan Society forthe Promotion of Science (JSPS, no. 16405001) and theAgriculture, Forestry and Fisheries Research Council(POMAL-STOPJELLY Project).

References

Arai MN (2001) Pelagic coelenterates and eutrophication: a re-view. Hydrobiologia 451: 69–87.

Barz K, Hirche HJ (2005) Seasonal development of scyphozoanmedusae and the predatory impact of Aurelia aurita on the zoo-plankton commuity in the Bornholm Basin (central Baltic Sea).Mar Biol 147: 465–476.

Brodeur RD, Sugisaki H, Hunt GL Jr (2002) Increases in jellyfishbiomass in the Bering Sea: implications for the ecosystem. MarEcol Prog Ser 233: 89–103.

Dawson MN (2003) Macro-morphological variation among cryp-tic species of the moon jellyfish, Aurelia (Cnidaria: Scyphozoa).Mar Biol 143: 369–379.

Dawson MN, Martin LE (2001) Geographic variation and ecologi-cal adaptation in Aurelia (Scyphozoa, Semaeostomeae): someimplications from molecular phylogenetics. Hydrobiologia 451:259–273.

Godhantaraman N, Uye S (2003) Geographical and seasonal vari-ations in taxonomic composition, abundance and biomass ofmicrozooplankton across a brackish-water lagoonal system ofJapan. J Plankton Res 25: 465–482.

Graham, WM, Kroutil RM (2001) Size-based prey selectivity anddietary shifts in the jellyfish, Aurelia aurita. J Plankton Res 23:67–74.

Hansson LJ, Moeslund O, Kiorboe T, Riisgård HU (2005) Clear-ance rates of jellyfish and their potential predation impact onzooplankton and fish larvae in a neritic ecosystem (LimfjordenDenmark). Mar Ecol Prog Ser 304: 117–131.

Ishii H, Tanaka F (2001) Food and feeding of Aurelia aurita inTokyo Bay with an analysis of stomach contents and a measure-ment of digestion time. Hydrobiologia 451: 311–320.

Kawahara M, Uye S, Ohtus K, Iizumi H (2006) Unusual popula-tion explosion of the giant jellyfish Nemopilema nomurai(Scyphozoa: Rhizostomeae) in East Asian waters. Mar EcolProg Ser 307: 161–173.

Kinoshita J, Hiromi J, Yamada Y (2006) Abundance and biomassof Scyphomedusae, Aurelia aurita and Chrysaora melanaster,and Ctenophora, Bolinopsis mikado, with estimation of theirfeeding impact on zooplankton in Tokyo Bay, Japan. JOceanogr 62: 607–615.

Koike F, Aizaki M, Seike Y, Akiba M, Okumura M, Fujinaga K(1999) Surface water exchange rate of the Honjo area in LakeNakaumi estimated from salinity change. Laguna 6: 19–25. (inJapanese with English abstract)

Kuwabara R (1969) Ecological study of medusa Aurelia auritaLamark—II. Distribution and biomass of Aurelia aurita in LakeHamana in summer 1965. Bull Mar Biol Stn Asamushi 13:193–199. (in Japanese with English abstract)

Kuwabara R, Sato S, Noguchi N (1969) Ecological studies on themedusae, Aurelia aurita Lamark—I. Distribution of Aurelia inthe north-eastern region of Tokyo Bay in summer of 1966 and1976. Bull Jpn Soc Sci Fish 35: 156–162. (in Japanese withEnglish abstract)

Lindahl O, Hernroth L (1983) Phyto-zooplankton community incoastal waters of western Sweden- an ecosystem off balance.Mar Ecol Prog Ser 10: 119–126.

Mills CE (2001) Jellyfish blooms: are populations increasingglobally in response to changing ocean conditions? Hydrobiolo-gia 451: 55–68.

Miyake H, Iwao K, Kakinuma Y (1997) Life history and environ-ment of Aurelia aurita. South Pacific Study 17: 273–285.

Möller H (1980) Scyphomedusae as predators and food competi-tors of larval fish. Meeresforschung 28: 90–100.

Olesen NJ (1995) Clearance potential of jellyfish Aurelia aurita,and predation impact on zooplankton in a shallow cove. MarEcol Prog Ser 63: 63–72.

Omori M, Ishii H, Fujinaga A (1995) Life history strategy of Au-relia aurita (Cnidaria, Scyphomedusae) and its impact on thezooplankton community of Tokyo Bay. ICES J Mar Sci 52:597–603.

Pagès F, González HE, Ramón M, Sobarzo M, Gili J-M (2001)Gelatinous zooplankton assemblages associated with watermasses in the Humboldt Current System, and potential preda-tory impact. Mar Ecol Prog Ser 210: 13–24.

Pearre Jr S (1982) Estimating prey preference by predators: use ofvarious indices, and a proposal of another based on c 2. Can JFish Aquat Sci 39: 914–923.

Purcell JE (2003) Predation on zooplankton by large jellyfish, Au-relia labiata, Cyanea capillata and Aequorea aequorea, in

Trophic relationship between common jellyfish Aurelia aurita and mesozooplankton 21

Prince William Sound, Alaska. Mar Ecol Prog Ser 246:137–152.

Purcell JE (2005) Climate effects on formation of jellyfish andctenophore blooms: a review. J Mar Biol Ass UK 85: 461–476.

Purcell JE, Arai MN (2001) Interactions of pelagic cnidarians andctenophores with fishes: a review. Hydrobiologia 451: 27–44.

Purcell JE, Sturdevant MV (2001) Prey selection and dietary over-lap among zooplanktivorous jellyfish and juvenile fishes inPrince William Sound, Alaska. Mar Ecol Prog Ser 210: 67–83.

Purcell JE, Uye S, Lo W-T (2007) Anthropogenic causes of jelly-fish blooms and their direct consequence for humans: a review.Mar Ecol Prog Ser 350: 153–174.

Rajagopal S, Nair KVK, Azariah J (1989) Some observations onthe problem of jelly fish ingress in a power station cooling sys-tem at Kalpakkam, east coast of India. Mahasagar 22: 151–158.

Schneider G, Behrends G (1994) Population dynamics and thetrophic role of Aurelia aurita medusae in Kiel Bight and west-ern Baltic. ICES J Mar Sci 51: 359–367.

Schneider G, Behrends G (1998) Top-down control in a neriticplankton system by Aurelia aurita medusae- a summary. Ophe-lia 48: 71–82.

Sullivan BK, Garcia JR, Klein-MacPhee G (1994) Prey selectionby the scyphomedusan predator Aurelia aurita. Mar Biol 121:335–341.

Uye S (1982) Length-weight relationships of important zooplank-ton from the Inland Sea of Japan. J Oceanogr 38: 149–158.

Uye S, Fujii N, Takeoka H (2003) Unusual aggregations of the

scyphomedusa Aurelia aurita in coastal waters along westernShikoku. Plankton Biol Ecol 50: 17–21.

Uye S, Nakai S, Aizaki M (2004) Potential use of extremely highbiomass and production rate of copepods in an enclosed brack-ish water body in Lake Nakaumi, Japan, for mass seed produc-tion of fishes. Zool Stud 43: 165–172.

Uye S, Shimauchi H (2005) Population biomass, feeding, respira-tion and growth rates, and carbon budget of the scyphomedusaAurelia aurita in the Inland Sea of Japan. J Plankton Res 27:237–248.

Uye S, Shimazu T (1997) Geographical and seasonal variations inabundance, biomass and estimated production rates of meso-and macrozooplankton in the Inland Sea of Japan. J Oceanogr53: 529–538.

Uye S, Ueta Y (2004) Recent increase of jellyfish populations andtheir nuisance to fisheries in the Inland Sea of Japan. Bull JpnSoc Fish Oceanogr 68: 9–19.

Watanabe T, Ishii H (2001) In situ estimation of ephyrae liberationfrom polyps of Aurelia aurita using settling plates in TokyoBay, Japan. Hydrobiologia 451: 247–258.

Yasuda T (1969) On the occurrence of the jellyfish ephyrae inTanjoura Bay, Fukui Prefecture. Aquaculture 17: 33–39. (inJapanese with English abstract)

Yasuda T (1983) Studies on the common jellyfish Aurelia aurita(Linne). Jpn Fish Resource Conserv Assoc, Tokyo, 189 pp. (inJapanese with English abstract)

22 C. HAN, M. KAWAHARA & S. UYE